What does this research mean for the field?

A framework combining autoencoders or principal component analysis with Deep Operator Networks (DeepONet) accurately predicts spatio-temporal responses in geotechnical problems such as soil consolidation, providing an efficient alternative to detailed deterministic numerical analysis. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

Develop and evaluate an efficient, data-driven reduced order modeling framework to predict complex non-linear geotechnical responses without high computational costs.

August 29, 2026Acta GeotechnicaOpen Access

A reduced order model framework suitable for geotechnical problems

View Full Paper

Ask AI

Bookmark

Authors

MOMao OuyangDurham University CACharles E. AugardeÉcole nationale des ponts et chaussées WCWilliam M. CoombsDurham University

Discussion

Member takes

Implication

Computational study demonstrates accurate spatio-temporal predictions of soil consolidation using autoencoders and DeepONet, indicating efficient alternatives to complex numerical simulations.

Key Points

Develop and evaluate an efficient, data-driven reduced order modeling framework to predict complex non-linear geotechnical responses without high computational costs.
Applied autoencoders and principal component analysis to reduce data dimensionality and extract latent representations from numerical models.
Integrated a Deep Operator Network (DeepONet) to capture and learn nonlinear system behaviors directly within the latent space.
Evaluated the framework on spatio-temporal response predictions during soil consolidation across multiple problem configurations.
Accurately predicted spatio-temporal responses across a range of soil consolidation specifications without detailed deterministic numerical analyses.
Established a versatile, computationally efficient modeling architecture capable of scaling to complex, large-scale geotechnical applications.

Cite This Study

Ouyang et al. (2026) studied this question.

synapsesocial.com/papers/6a9299338e5d7d1fc0c10f4a https://doi.org/10.1007/s11440-026-03221-0

View Full Paper

Ask AI

Bookmark

Implication

Computational study demonstrates accurate spatio-temporal predictions of soil consolidation using autoencoders and DeepONet, indicating efficient alternatives to complex numerical simulations.

A reduced order model framework suitable for geotechnical problems

Discussion

Member takes

Implication

Key Points

Cite This Study

A reduced order model framework suitable for geotechnical problems

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider